IFIP Open Access Digital Library
Not a member yet
22614 research outputs found
Sort by
Controlling Popularity Bias in Sequential Recommendation Models
No full textPart 2: Recommendation/ClassificationInternational audienceSociety today has witnessed a rapid increase in the prevalence of social media and popular content delivery websites such as Netflix and Spotify. As a result, through the way that people consume news and media, we are transitioning from a static media delivery model to a dynamic, personalized system which many are adapting and even enjoying the resulted changes. Personalized recommendations are mostly made with the help of the machine learning models embedded in recommendation systems (RS). It is important to examine these systems that are becoming very relevant in our daily lives, and to identify how we can either control or at least mitigate any harmful effects caused by these new systems. One of those prominent issues within RS is popularity bias, which stems from the nature of machine learning, which often prioritizes popularity over novelty. That is, machine learning methods tend to prioritizes being most correct rather than trying to find a truly fitting recommendation. Popularity bias is a main cause of echo chambers within the current media landscape, which unfortunately has led to less critical thinking and more divisiveness our communities. To counter this issue, we present a novel methodology that combines two existing methods proposed for sequential RS, aiming to reduce popularity bias without overly punish popular items. The result is a weighted RS that shows the promising ability to control the amount of popularity bias. The proposed system may be applied to multi-model hybrid RS, which would achieve personalized recommendations that support individual weighted preference between popularity and novelty
Towards Atomicity and Composability in Cross-Chain NFTs
No full textPart 2: Crypto ApplicationInternational audienceNon-Fungible Tokens (NFTs) have created new and interesting ways to own digital assets. Our work defines non-fungible tokens as a pair and addresses challenges in cross-chain NFT transfers, particularly on interoperability between siloed blockchains. We develop a methodology and propose a Greed Factor to prioritize various hashing methods for detecting duplicates. Our practical approach marks an advancement over existing methods by using an analysis of over 400,000 real-world NFTs across 50 collections, which is easily applicable to existing bridges. The experimental results show the limitations of current implementations, and we offer insights to strengthen their functionality and foster the future growth of NFTs
From Play to Profession: A Serious Game to Raise Awareness on Digital Forensics
No full textPart 6: Security User StudiesInternational audienceWith the increasing digitization and interconnectivity of organizations, the frequency of cyberattacks is rising. These attacks have serious consequences for data security and critical infrastructure. However, the persistent lack of cybersecurity specialists represents a major challenge for organizations. One approach to address this issue is to use serious games for career orientation in schools to steer people interested in security or digital forensics (DF) at an early age. Thus, we introduce a serious game called Digital Detectives, designed to familiarize students and young IT professionals with DF practices. Players adopt the role of forensic investigators tasked with probing a cyberattack on a company and acquire hands-on expertise by gathering evidence, identifying perpetrators, and recovering encrypted data. We evaluated the serious game with 102 young students in a pre- and post-test. We find that our serious game has a great game experience and significantly increases learning outcomes and awareness of DF. Our results indicate that the games’ learning outcomes and game experience are strongly correlated to increased awareness of DF. In sum, the serious game is an effective means to aid students in career orientation and raise awareness of DF
DT-Anon: Decision Tree Target-Driven Anonymization
No full textPart 3: PrivacyInternational audienceMore and more scenarios rely today on data analysis of massive amount of data, possibly contributed from multiple parties (data controllers). Data may, however, contain information that is sensitive or that should be protected (e.g., since it exposes identities of the data subjects) and cannot simply be freely shared and used for analysis. Business rules, restrictions from individuals (data subjects to which data refer), as well as privacy regulations demand data to be sanitized before being released or shared with others. Unfortunately, such protection typically comes with a loss of utility of the released data, impacting the performance of the analytics tasks to be executed.In this paper, we present DT-Anon, a target-driven anonymization approach that aims at protecting (anonymizing) data while preserving as much as possible the capability of a classification task operating downstream to learn from the anonymized data. The basic idea of our approach is to perform the anonymization process on partitions produced by a decision tree driven by the target of the classification task. Each partition is then independently anonymized, to limit the impact of anonymization on the attributes and values that work as predictors for the target of the classification task. Our experimental evaluation confirms the effectiveness of the approach
E-Participation Without Democracy: Understanding Variation in Digital Engagement in Non-democracies
No full textInternational audienceThe variation in E-participation adoption and obstruction among non-democratic regimes is not sufficiently understood in earlier research. We attribute this to a lack of conceptual instruments for systematically studying the regime attributes of non-democratic states. Inspired by the work of Linz and Stepan (1), we demonstrate how a more fine-grained and multi-dimensional taxonomy of non-democratic regimes could differentiate between regime behaviours. Based on this categorisation, we further formulate expectations regarding four dimensions of E-participation in different types of non-democratic regimes. We argue that the proposed regime categorisation and identified expectations can form a basis for more nuanced comparative research on E-participation in non-democratic states
Generative AI-Augmented Decision-Making for Business Information Systems
No full textInternational audienceThe integration of Generative Artificial Intelligence (GAI) in decision-making has ushered in a new era of opportunities and challenges for organizations. However, due to the way GAI algorithms work, the propagation of social biases and the lack of transparency on how they use data raises concerns about the autonomy and control of human decision-making powered by such systems. In this experimental research paper, we contrast the answers that ChatGPT, a popular GAI tool, can give us in the context of decision-making with what we know about such processes from the management and business information systems literature. Our findings suggest that GAI can facilitate the organization of information and options for making decisions. However, without a moral and ethical stance, the responsibility for the decisions remains with the human actor. Suggesting a collaborative approach between humans and GAI, we reflect on the changes to learning and adaptation patterns that need to happen on both sides to reform the way we make GAI-augmented decisions
Impact of a Set of Factors on Order Lead Time: A Case Study of an Apparel Company
No full textPart 6: Uncertainty and Collaboration in Supply ChainInternational audienceApparel supply chain is facing challenges due to recent unexpected market turbulences. This might affect the order lead time with consequences on customers’ satisfaction. It is therefore important to have more control over the supply chain to be more resilient and to respond to market changes. In this context, the distances that products must travel from the production plants can be very significant, influencing order lead times. Consequently, in order to quantitatively measure the effect of distance, as well as other operation independent factors, on order lead time, an empirical model is presented. The data is collected from orders placed by a clothing company that sells its products internationally and outsources production. The results show a significant relationship between lead time, distance or proximity, quantity actually shipped, order cost and product category ordered
Integrating social interaction within senselife framework
No full textPart 1: Empowering Vulnerable Populations Well-being through Collaborative NetworksInternational audienceAs the global elderly population continues to grow, social isolation emerges as a critical factor contributing to the incidence of frailty. This paper explores the integration of social interaction functionalities within the Senselife framework, a service recommendation platform designed for frailty prevention in older adults. We propose enhancements to Senselife that facilitate communitybuilding and social engagement through technology-driven interactions. By leveraging user-centered design, the paper discusses how enhanced social features can significantly improve the efficacy of our frailty prevention strategies, offering a holistic approach to elderly care. Our methodology includes the development of social interaction modules that encourage active participation and connectivity among elderly users, ultimately aiming to enhance their quality of life and reduce the risks associated with social isolation
A Second-Order Adaptive Network Model for Political Opinion Dynamics
No full textPart 3: Data Mining/ModelingInternational audienceThis paper introduces a second-order adaptive network model for simulating political opinion dynamics, considering cognitive, affective, and social factors. The model, grounded in political psychology and communication theories, illustrates how individuals’ opinions evolve in response to external stimuli such as political parties and media. It also explores the impact of individuals’ reasoning abilities and initial viewpoints on their rationality and cognitive flexibility. Through simulation experiments, the paper demonstrates the model’s capacity to generate realistic outcomes such as homophily and polarization phenomena. It further discusses the model’s implications for mitigating misinformation spread and reducing polarization in political opinion dynamics, identifying key influencing factors and potential interventions. The paper contributes to computational politics by offering an innovative approach to modeling individual and collective opinion formation processes, acknowledging the complexity and adaptivity of cognitive, affective, and social dynamics
Artificial Intelligence Applications and Innovations: 20th IFIP WG 12.5 International Conference, AIAI 2024, Corfu, Greece, June 27–30, 2024, Proceedings, Part II
No full textInternational audienceBook Front Matter of AICT 71